Scoring Recommendations and Explanations with a Probabilistic User Model

1. A computer system to score recommendations for potential purchase by a customer and provide candidate recommendations, the computer system comprising: a processor: and a memory, coupled to the processor, storing code that is executable by the processor to configure the computer system into a machine to: receive recommendation context digital data of the customer; process the recommendation context digital data, by evaluating which rules match the recommendation context digital data and removing evaluated rules with a filter if evaluated values do not meet a recommendation strength threshold, to identify a plurality of candidate recommendations that match the recommendation context digital data, where each candidate recommendation recommends at least one recommended item; determine a score utilizable by the computer system for each candidate recommendation by item applying a purchase decision model to score an electiveness of each candidate recommendation by comparing a recommendation outcome against a non-recommendation outcome; and rank the plurality of candidate recommendations using the score for each candidate recommendation to identify at least a highest ranking candidate recommendation; select the candidate recommendation having the highest score for issuance to the customer; and provide digital data representing the candidate recommendation to a computer system accessible to the customer.

2. The computer system of claim 1 wherein the code is further executable by the processor of the computer system to configure the computer system into the machine for performing: selecting a message selector that uses a probabilistic use case model to choose, for each candidate recommendation, a selling message, thereby generating a recommendation, message pair for each candidate recommendation.

3. The computer system of claim 2 wherein the code is further executable by the processor of the computer system to configure the computer system into the machine for using the probabilistic purchase decision model to choose a candidate recommendation based on the recommendation, message pair that maximizes boost in expected margin resulting from the candidate recommendation being issued, as compared to the candidate recommendation not being issued.

4. The computer system of claim 1 wherein the code is further executable by the processor of the computer system to configure the computer system into the machine that implements the probabilistic purchase decision model as a Bayesian network.

5. The computer system of claim 1 wherein the code is further executable by the processor of the computer system to configure the computer system into the machine for: generating a selling point message for each candidate recommendation by: identifying a plurality of selling point messages corresponding to the candidate recommendation, where each selling point messages is targeted to a predetermined user case profile; identifying a first user case profile most likely matches the customer; and selecting a first selling point message from the plurality of selling point messages that is targeted to the first user case profile for use with the candidate recommendation.

6. The computer system of claim 1 wherein determining a score for each candidate recommendation by applying a probabilistic purchase decision model to score an effectiveness of each candidate recommendation by comparing a recommendation outcome against a non-recommendation outcome comprises: using a Bayesian network to calculate a first purchase probability measure for a candidate recommendation item, given a relevance measure for the item, an exposure measure for the item if the candidate recommendation is displayed and a clarity measure for the item if the candidate recommendation is displayed; using a Bayesian network to calculate a second purchase probability for the candidate recommendation item, given a relevance measure for the item, an exposure measure for the item if the candidate recommendation is not displayed and a clarity measure for the item if the candidate recommendation is not displayed; subtracting the second purchase probability measure from the first purchase probability measure to form a first difference value; and multiplying the first difference value by a margin value for the candidate recommendation item.

7. A non-transitory, computer readable medium comprising code stored therein, wherein the code is executable by a processor of a computer system to enable the computer system to score recommendations for potential purchase by a customer and provide candidate recommendations by configuring the computer system into a machine to perform: receiving recommendation context digital data of the customer; processing the recommendation context digital data, by evaluating which rules match the recommendation context digital data and removing evaluated rules with a filter if evaluated values do not meet a recommendation strength threshold, to identify a plurality of candidate recommendations that match the recommendation context digital data in accordance with association rules not removed, where each candidate recommendation recommends at least one recommended item; determining a computer system useable score for each candidate recommendation by applying a purchase decision model to score an effectiveness of each candidate recommendation by comparing a recommendation outcome against a non-recommendation outcome; ranking the plurality of candidate recommendations using the score for each candidate recommendation to identify at least a highest ranking candidate recommendation; selecting the candidate recommendation having the highest score for issuance to the customer; and providing digital data representing the candidate recommendation to a computer system accessible to the customer.

8. The non-transitory, computer readable medium of claim 7 wherein the code is further executable by the processor of the computer system to configure the computer system into the machine for performing: selecting a message selector that uses a probabilistic use case model to choose, for each candidate recommendation, a selling message, thereby generating a recommendation, message pair for each candidate recommendation.

9. The non-transitory, computer readable medium of claim 8 wherein the code is further executable by the processor of the computer system to configure the computer system into the machine for using the probabilistic purchase decision model to choose a candidate recommendation based on the recommendation, message pair that maximizes boost in expected margin resulting from the candidate recommendation being issued, as compared to the candidate recommendation not being issued.

10. The non-transitory, computer readable medium of claim 7 wherein the code is further executable by the processor of the computer system to configure the computer system into the machine that implements the probabilistic purchase decision model as a Bayesian network.

11. The non-transitory, computer readable medium of claim 7 wherein the code is further executable by the processor of the computer system to configure the computer system into the machine for: generating a selling point message for each candidate recommendation by: identifying a plurality of selling point messages corresponding to the candidate recommendation, where each selling point messages is targeted to a predetermined user case profile; identifying a first user case profile most likely matches the customer; and selecting a first selling point message from the plurality of selling point messages that is targeted to the first user case profile for use with the candidate recommendation.

12. The non-transitory, computer readable medium of claim 7 wherein determining a score for each candidate recommendation by applying a probabilistic purchase decision model to score an effectiveness of each candidate recommendation by comparing a recommendation outcome against a non-recommendation outcome comprises: using a Bayesian network to calculate a first purchase probability measure for a candidate recommendation item, given a relevance measure for the item, an exposure measure for the item if the candidate recommendation is displayed and a clarity measure for the item if the candidate recommendation is displayed; using a Bayesian network to calculate a second purchase probability for the candidate recommendation item, given a relevance measure for the item, an exposure measure for the item if the candidate recommendation is not displayed and a clarity measure for the item if the candidate recommendation is not displayed; subtracting the second purchase probability measure from the first purchase probability measure to form a first difference value; and multiplying the first difference value by a margin value for the candidate recommendation item.